© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 33243
    Prev
    Next
    INDEX
    Explanations

    references to sexuality and provocative content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    endid
    -0.16
     Hairst
    -0.16
    yntax
    -0.15
    nict
    -0.15
    ocide
    -0.15
     frags
    -0.15
    osemite
    -0.15
    pollo
    -0.14
    BOSE
    -0.14
    CID
    -0.14
    POSITIVE LOGITS
     scandal
    0.26
     explicit
    0.24
     shocked
    0.23
     sexual
    0.23
     taboo
    0.22
     tit
    0.22
     controversial
    0.22
     controversy
    0.22
     shock
    0.22
     voy
    0.21
    Activations Density 0.069%

    No Known Activations